148 92 77MB
English Pages 232 [316] Year 2023
LANGUAGE LEARNING STRATEGIES AROUND THE WORLD: CROSS-CULTURAL PERSPECTIVES
TECHNICAL REPORT #13
LANGUAGE LEARNING STRATEGIES AROUND THE WORLD: CROSS-CULTURAL PERSPECTIVES edited
H
by R E B E C C A
OXFORD
SECOND LANGUAGE TEACHING & CURRICULUM CENTER U n i v e r s i t y
of
H a w a i 'i
at
M ä n o a
© 1996 Second Language T e a c h i n g & Curriculum C e n t e r University of Hawai'i All rights reserved Manufactured in t h e U n i t e d States of America
Funds for the publication of this technical report were provided in part by a grant to t h e University of Hawai'i under t h e Language Resource C e n t e r s Program of t h e U S Department of Education.
ISBN 0 - 8 2 4 8 - 1 9 1 0 - 1
iaoT T h e paper used in this publication meets the minimum requirements of t h e A m e r i c a n N a t i o n a l S t a n d a r d for Information S c i e n c e s - P e r m a n e n c e of Paper for Printed Library Materials. A N S I Z39.48-1984 Book design by Deborah M astersoti Distributed by University of Hawai'i Press Order Department 2840 Kolowalu Street Honolulu, HI 96822
ABOUT THE NATIONAL FOREIGN LANGUAGE RESOURCE CENTER
THE SECOND LANGUAGE TEACHING AND CURRICULUM CENTER of the U n i v e r s i t y of H a w a i ' i is a unit of the C o l l e g e of L a n g u a g e s , Linguistics, a n d Literature. U n d e r a grant from the U S D e p a r t m e n t of E d u c a t i o n , the C e n t e r has since 1990 served as a N a t i o n a l Foreign L a n g u a g e Resource C e n t e r ( N F L R C ) . T h e general direction of the Resource C e n t e r is set by a national advisory board. T h e C e n t e r conducts research, develops materials, and trains language professionals with the goal of improving foreign language instruction in the U n i t e d States. T h e C e n t e r publishes research reports and t e a c h i n g materials; it also s p o n s o r s a s u m m e r i n t e n s i v e t e a c h e r training institute. For a d d i t i o n a l i n f o r m a t i o n a b o u t C e n t e r programs, write:
Dr. Richard S c h m i d t , Director N a t i o n a l Foreign Language Resource C e n t e r East-West R o a d , Bldg. 1, R m . 6 A University of Hawai'i H o n o l u l u , HI 9 6 8 2 2 or visit our W e b site: http://www.lll.hawaii.edu/nflrc
•
V
N F L R C ADVISORY BOARD Kathleen Bardovi-Harlig Center for English Language Teaching Indiana University John Clark Defense Language Institute Monterey, California James Pusack Project for International Communication Studies (PICS) University of Iowa Ronald Walton National Foreign Language Center Washington, DC Representatives of other funded NFLRCs
Vi
•
CONTENTS Preface:
PART I:
Why is Culture Important for Language Learning Strategies? Rebecca L. Oxford
ix
WHAT DOES STRATEGY ASSESSMENT TELL US IN VARIOUS CULTURES AND LANGUAGES?
Chapter 1:
Increasing Metacognitive Awareness in the L2 Classroom by Using Think-Aloud Protocols and Other Verbal Report Formats Neil J. Anderson & Laurens Vandergrifc
3
Chapter 2:
Telling Their Stories: Language Students Use Diaries and Recollection Rebecca L. Oxford, Roberta Z. Lavine, Gregory Felkins, Mary Evelyn Hollaway, & Amany Saleh
19
Relationship Between Language Learning Strategies and Israeli Versus Russian Cultural-Educational Factors Adina Levine, Thea Reves, & Betty Lou Leaver
35
Cross-Cultural Comparisons of Language Learning Strategies in the People's Republic of China and Other Countries David A. Bedell & Rebecca L. Oxford
47
Chapter 5: Learning Strategies and Other Predictors of ESL Proficiency Among Afrikaans Speakers in South Africa Carisma Drey er & Rebecca L. Oxford
61
Chapter 3:
Chapter 4:
Chapter 6:
The Influence of Gender and Motivation on EFL Learning Strategy Use in Jordan Cora Kaylani
Chapter 7:
75
A Synthesis of Approaches to Assessing Language Learning Strategies Andrew D. Cohen & Kimberly Scott
89
PART II: HOW CAN WE IMPROVE STRATEGY USE AROUND THE WORLD? Chapter 8:
The Conceptual Shift to Learner-Centered Classrooms: Increasing Teacher and Student Strategic Awareness Martha Nyikos
109
•
vii
Chapter 9:
How Print Materials Provide Strategy Instruction Maaike Hajer, Theun Meestringa, Young Ye Park, & Rebecca L. Oxford
119
Chapter 10: Unobtrusive Computerized Observation of Compensation Strategies for Writing to Determine the Effectiveness of Strategy Instruction Carol Ann Baily
141
Chapter 11: Using Multimedia for Learner Strategy Instruction Joan Rubin
151
Chapter 12: University-Level Studies Using Strategy Instruction to Improve Speaking Ability in Egypt and Japan El Sayed Dadour & Jill Robbins
157
Chapter 13: Implementing the Cognitive Academic Language Learning Approach (CALLA) Anna Uhl Chamot & J. Michael O'Malley
167
Chapter 14: Methods for Teaching Learning Strategies in the Foreign Language Classroom Anna Uhl Chamot, Sarah Bamhardt, Pamela El-Dinary, & Jill Robbins
175
Chapter 15: The Foresee Approach for ESL Strategy Instruction in an Academic-Proficiency Context Richard Kidd & Brenda Marquardson
189
Chapter 16: Effective Awareness-Raising in Language Learning Strategy Instruction Nae-Dong Yang
205
Chapter 17: A Two-Phase Study Involving Consciousness Raising and Strategy Use for Foreign Language Learners Jeffra Flaitz & Carine Feyten
211
Chapter 18: A Synthesis of Strategy Instruction for Language Learners Rebecca L. Oxford & Betty Lou Leaver
227
Afterword: What Have We Learned About Language Learning Strategies Around the World? Rebecca L. Oxford
247
References
251
About the Authors and How to Contact Them
283
viii
•
Rebecca L. Oxford University of Alabama, US A
PREFACE WHY IS CULTURE IMPORTANT FOR LANGUAGE LEARNING STRATEGIES? Culture is the social cement of all human relationships. -TomScovel, 1991
This book, which focuses o n cross-cultural use of language learning strategies, is useful to foreign language teachers around t h e world w h o have students from just o n e culture. It is also helpful for second language teachers w h o have more t h a n one culture represented in their classes. Language researchers and language program administrators would also benefit from a greater and more practical understanding of cross-cultural similarities and differences in learning strategy assessment and instruction. Finally, advanced language learners are likely to gain from a greater comprehension of the cultural dynamics of strategy use. Culture and its influence on language learning strategies are t h e keystones of this preface and indeed of this whole book. This preface defines culture and cross-culturalism, defines learning strategies and explains cultural influences on those strategies, and displays the organization and special features of this book.
CULTURE AND CROSS-CULTURALISM Culture (relating to patterns of living) refers to the individual's role in the unending kaleidoscope of life situations of every kind and t h e rules or models for attitudes and conduct in them. By reference to these models, all hum a n beings, from infancy onward, justify t h e world to themselves as best they can, associate with those around them, and relate to t h e social order to which they are attached... W h a t is important in culture... is what o n e is expected to think, believe, say, do, eat, wear, pay, ensure, resent, h o n o r , laugh at, fight for, and worship, in typical life situations... (Brooks, 1968, pp. 218-221) T h e famous metaphor of the "cultural iceberg" (Hall & Hall, 1990; Oxford, 1995) indicates that many aspects of culture, such as certain beliefs, perceptions, and values, are below the surface of consciousness (in t h e submerged part of t h e iceberg). O t h e r aspects of culture, like clothing and TV-watching habits, are in the conscious
•
ix
area (above the waterline). T h e less conscious cultural aspects often influence how people learn languages. Research by Yang ( 1 9 9 2 b ) suggests that culture clearly includes beliefs, perceptions, and values which affect language learning, including general learning styles (visual, auditory, hands-on; intuitive, sensing; global, analytic; see Reid, 1 9 9 5 ) and specific learning strategies (the particular behaviors and steps learners use to improve their learning such as note-taking, finding conversation partners, and analyzing words). Oxford, Hollaway, and Horton-Murillo ( 1 9 9 2 , p. 4 4 1 ) emphasize, " A l t h o u g h culture is not the single determinant, and although many other influences intervene, culture often does play a significant role in the learning styles [and strategies]... adopted by many participants in the culture." T h e importance of culture is reflected in the concept of "situated cognition," which holds that the setting and the activity in which knowledge is developed are not separable from learning, nor are they neutral; they are an integral part of the learning (Brown, Collins, & Duguid, 1989; Geertz, 1983; Lave, 1988; Lave &. Wenger, 1991; Rogoff & Lave, 1984; Suchman, 1987). Thus, in the foreign or second language classroom, the activities and cultural influences cannot be separated from what is learned. Language learning is fully situated within a given cultural context. T h e student becomes enculturated (apprenticed into a particular learning culture that in many ways reflects the general culture) through classroom activities and through the modeling and coaching of the teacher and many others (Lave, 1988; Rogoff & Lave, 1 9 8 4 ) . Rather than just the teacher/learner dyad, there exists "a richly diverse field of essential actors and, with it, other forms of relationships of participation" (Lave &. Wenger, 1991, p. 56). In this view, learning is never a mere process of transmission or transfer but is instead nothing less than a process of transformation. Cross-Culturalism (Banks & Banks, 1993; Batchelder &. Warner, 1977; Gaston, 1984; Luce & S m i t h , 1987; Oxford, 1995; Putsch, 1986; Seelye, 1987; Weeks, Pedersen, & Brislin, 1 9 7 7 ) deals with a dynamic system of understandings across cultures or subcultures. Comprehending cross-cultural similarities and differences involves carefully considering crucial aspects of culture, such as concepts of time, personal space, body language, worship, relationships, hatred, prejudice, love, and respect — and, as shown in this book, language learning strategies within a particular culture and across cultures.
LANGUAGE LEARNING STRATEGIES IN THE CULTURAL CONTEXT Strategies are the tools for active, self-directed involvement that is necessary for developing communicative ability ( W e n d e n &. Rubin, 1987; O'Malley &. Chamot, 1 9 9 0 ) . Strategies are not a single event, but rather a creative sequence of events that learners actively use. Hundreds of foreign and second learning strategies exist. For example, Bob seeks out conversation partners. Zoltan groups words to be learned and then labels each group. Marlen gives herself encouragement through positive
X
•
self-talk before getting up to give a speech in the target language. Louelle uses gestures to communicate in the classroom when the words do not come to mind. Mariam learns words by breaking them down into their components. Sayed draws "semantic maps" with lines and arrows pictorially showing the linkages between new words according to their meaning. Deena finds an American pen pal. Darlene consciously uses guessing while reading as many books as she can in the second language. L2 learning strategies like these are very important, because research has repeatedly shown that the conscious, "tailored" use of these strategies is related to language achievement and proficiency. Learning strategy investigations within and outside the language field have shown that effective learners actively associate new information with existing information in long-term memory, building increasingly intricate and differentiated mental structures, or schemata. The use of well-chosen strategies distinguishes experts from novices in many learning areas. Successful learners often use metacognitive (i.e., "beyond the cognitive") strategies such as organizing, evaluating, and planning their learning. These are sometimes viewed as the learner's own personal "executive control" over his or her own learning. Use of these behaviors — along with cognitive strategies like analyzing, reasoning, transferring information, taking notes, and summarizing — might be considered part of any definition of truly effective learning (Brown, Bransford, Ferrara, & Campione, 1983). Additionally, competent learners often use compensation strategies such as guessing or inferring and memory strategies such as grouping and structured reviewing — all of which have been included as cognitive strategies by most researchers. Research results have shown that some of the best learners use affective and social strategies to control their emotional state, to keep themselves motivated and on-task, and to get help when they need it (McCombs, 1982, 1988; Dansereau, 1985). Yet many students (and their teachers) are largely unaware of the potential of affective and social strategies. Cultural background affects strategy choice. Because of Hispanics' global and field dependent style preference (Reid, 1995), many Hispanic ESL/EFL students choose particular learning strategies, such as predicting, inferring (guessing from context), avoiding details, working with others rather than alone, and basing judgments on personal relationships rather than logic. In contrast, many Japanese ESL/EFL students reflectively use analytic strategies aimed at precision and accuracy, search for small details, work alone, and base judgments more on logic than on personal interactions. Cultures that encourage concrete-sequential learning styles (such as those of Korea or some Arabic-speaking countries) often produce widespread use of rote memorization strategies, while more flexible strategies (though not always higher order thinking strategies) and a more facilitative view of teachers are often found among North Americans. Extroverted learning styles, such as those of many Hispanics and Arabic speakers, are related to the use of social strategies for learning (Harshbarger et al., 1986; Willing, 1988). In contrast, according to Harshbarger et al. (1986), many introverted Asian students display strategies for working alone. These are just a few examples of cultural influences on learning strategy selection and use. In addition, gender differences in language learning strategy use have been
•
xi
rampant in the research (Oxford, 1993a, 1993b), and surely culture is at least one feasible, if partial, explanation. Attitudes toward authority, beliefs about how difficult (or easy) it is to learn a language, ideas about the importance of "the text" and about memorizing it, and concepts about whether personal expression and creativity are allowed — all these are cultural issues that affect the use of language learning strategies. Culture also influences strategy assessment, the measurement of strategy use. T e c h niques often used for assessing students' language learning strategies include informal observation, formal observational rating scales, informal or formal interviews, group discussions, think-aloud procedures, language learning diaries, dialogue journals between student and teacher, open-ended narrative-type surveys, and structured questionnaires of strategy frequency (see, e.g., C o h e n , 1987a). Learners in some cultures might feel too vulnerable in a think-aloud strategy assessment situation and might prefer a strategy assessment questionnaire. Likewise, culture influences strategy instruction. Students in certain cultures might like large and small group discussions with teacher input, while students elsewhere might prefer to improve their strategies independently through self-directed workbooks. T h e following is an authentic example of how culture influences strategy use and attitudes toward learner empowerment. It was written by English teacher Milagros Flores in Venezuela. N o t i c e the great surprise registered by some students and the skepticism of others when faced with the concept that they themselves could active!y do something to enhance their learning: W h e n e v e r a strategy is used by a student in class, I stop and alert the whole class of that strategy and its effectiveness. I also try to use activities which would encourage students to use strategies, for example information gap activities, problem solving, and guessing the meaning from context. In my case after I arrived from the [strategy course for teachers], my first day of class I told my students about this new finding [learning strategies can help students learn more effectively] and I let them know most of the theory. But I realize it is big change for Latin people, I mean, it is going to involve a lot of time for them to get used to this new finding. My students' reaction varied from surprised to skeptical or mocking. Because our students are not used to being an active part of the learning process, we have to be persistent in the discussion of the strategies and make students aware of them every time they obtain good results. In this culture, passivity was the norm in the learning process. In other cultures the sense of passivity in learning might not be so pronounced, and personal action and power might be felt more strongly. Each culture has its approach to learning and thus to learning strategies, and therefore no single formula for assessing and instructing learning strategies exists.
xii
•
HOW THIS BOOK IS ORGANIZED Part I concerns strategy assessment in various cultures and with different languages, and Part II delves deeply into strategy instruction as it takes place with students of contrasting cultures and languages. Chapter 1, by Anderson and Vandergrift, focuses on verbal reports, especially the well-known think-aloud procedure, and presents a major Canadian illustration highlighting learners of French as a second language. Oxford, Lavine, Felkins, Hollaway, and Saleh in chapter 2 discuss the use of diary studies and recollective studies and give examples with a variety of languages and countries. Using a variety of assessment techniques Levine, Reves, and Leaver demonstrate in chapter 3 how ex-Soviets who have been in Israel for a short time and those who have lived in Israel for decades differ in their culturaleducational learning views and their learning strategy use. Bedell and Oxford in chapter 4 review 36 studies of language learning strategy use in many different cultures, and then they center on an intensive strategy study in the People's Republic of C h i n a . Chapter 5 by Dreyer and Oxford is a view of South Africa, focusing on personality factors and learning strategies among Afrikaans-speaking learners of ESL. Kaylani, author of chapter 6, studies motivation and gender as related to learning strategy use in Jordan. In chapter 7, C o h e n and Scott give a synthesis of i n t e r n a t i o n a l l y used strategy assessment options, g i v i n g a d v a n t a g e s a n d disadvantages of each, as a fitting close to Part I. Part II, with its focus on strategy instruction, starts with chapter 8 by Nyikos, who explains the conceptual shift that is essential when the teacher wants to create a learner-centered classroom (a necessity for improving students' learning strategies). Hajer, Meestringa, Park, and Oxford show in chapter 9 the power of print materials in strategy instruction, using illustrations from the UK, the U S , and the Netherlands. Baily's chapter 10 proves that the computer can be a vehicle for assessing/tracking language learning strategies and for providing certain forms of strategy instruction. In chapter 11, Rubin continues the topic of media by introducing her Language Learning Strategy Program, which provides multimedia strategy instruction regarding multiple target languages, such as Russian, Korean, Spanish, and 17 others. Chapter 12 by Dadour and Robbins is a demonstration of the effectiveness of detailed, systematic strategy instruction for EFL oral communication skills in Egypt and in Japan. Chapter 13 details the C A L L A model in a discussion by Chamot and O'Malley. Many of the principles of that model are applied to Japanese, Spanish, and Russian learning in chapter 14, written by Chamot, Barnhardt, El-Dinary, and Robbins. Kidd and Marquardson propose a well-articulated ESL strategy instruction model from Canada, a model known as Foresee and based somewhat on C A L L A but with very specific revisions (chapter 15). In chapter 16, Yang shows how a strategy instruction program designed with group strategy interviews and discussions can affect the frequency and variety of strategy use among Taiwanese university students. Flaitz and Feyten describe a vibrant but short strategy instruction i n t e r v e n t i o n and show that it can, in optimal circumstances, raise both strategic awareness and language performance. Chapter 18 by Oxford and Leaver synthesizes what we know at this time about strategy
•
xiii
instruction around the world and refers to several key factors, such as consciousness, resource use, and degree of integration into regular classwork. The afterword brings together the key points of Parts I and II.
SPECIAL FEATURES OF THIS BOOK This volume has a number of special characteristics not found in many other books on language learning strategies.
xiv
•
This is the first book that looks closely at the nature of cultural effects on language learning strategy use, assessment, and instruction. Areas cited in this book for their language learning strategy involvement (and often for cultural influences) are Belarus, Canada, Denmark, Egypt, France, Israel, Japan, Jordan, the Netherlands, the Philippines, Puerto Rico, the People's Republic of China, the Republic of China (Taiwan), Russia, Saudi Arabia, Thailand, the Ukraine, the U S , the UK, and many more.
•
Unlike many other strategy books, this book discusses a wide range of cuiturally appropriate and practical ways to conduct strategy assessment and instruction.
•
This is the first volume that shows how language learning strategies are employed, assessed, and taught with a large variety of native and second/foreign languages. In this book, a sampling of the languages specifically affected by language learning strategies includes, among others: Danish, Dutch, English (as a second language and as a foreign language), French, German, Hebrew, Japanese, Russian, and Spanish.
•
This book contains chapters by most of the well-known researchers in the field of language learning strategies and also introduces some young, highly promising strategy researchers.
•
This book contains new, unpublished information about language learning strategies; for instance, the volume describes recent strategy assessment and instructional projects, explores new links between strategy use and language performance, discusses computerized strategy assessment, highlights the value of print media for self-directed strategy instruction, and shows that a small amount of strategy instruction can sometimes have a surprisingly significant effect on language achievement.
•
T h e volume is easier to read than most edited books in the language learning/teaching field. It has been carefully edited by a single individual, who aimed at eliminating major redundancies, creating smoothness and uniform quality across chapters, avoiding jargon, providing the clearest and most powerful presentation possible, offering syntheses at the end of each part of the book, and gathering references at the end of the volume.
•
As you read this book, consider the questions you want to have answered about language learning strategies around the world. Search for what you personally need to know. At the same time, keep an open mind for new questions and ideas you have not even considered. Enjoy this cross-cultural exploration of language learning strategies.
•
XV
PART I: WHAT DOES STRATEGY ASSESSMENT TELL US IN VARIOUS CULTURES AND LANGUAGES?
Neil J. Anderson Ohio University, USA Laurens Vandergrift University of Ottawa, Ontario, Canada
CHAPTER 1 INCREASING METACOGNITIVE AWARENESS IN THE L2 CLASSROOM BY USING THINK-ALOUD PROTOCOLS AND OTHER VERBAL REPORT FORMATS Speak out: What is it thou hast heard, or seen? -Lord Tennyson
SUMMARY Think-aloud protocols (and other verbal report formats) constitute a main technique for assessing students' learning strategies. This technique involves asking students to express their thoughts or strategies while they do a language task or soon thereafter. This chapter offers historical and conceptual background information and helpful guidelines regarding the use of such protocols. It also provides numerous examples, including a major illustration from Canada.
INTRODUCTION Think-aloud protocols, along with related retrospective verbal report protocols, have been used in many second language (L2) research efforts to uncover the mental processes or strategies that second language learners use while engaged in language learning tasks. Such protocols allow "insight into the dynamic and interactive nature" of the language learning process (MacLean and d'Anglejan, 1986, p. 814). These protocols have helped redirect attention from a focus on products of language use to a focus on the ongoing process, which includes attending to language input, arriving at spoken utterances, processing a text, generating a text, and learning and retrieving vocabulary. Using think-aloud protocols and other verbal report formats is a beneficial metacognitive activity and helps students become more aware of the options available to them in understanding language and being a better language learner.
Anderson, N. J., & Vandergrift, L. (1996). Increasing metacognitive awareness in the L2 classroom by using think-aloud protocols and other verbal report formats. In Rebecca L. Oxford (Ed.), Language learning strategies around the world: Cross'cultural perspectives. (Technical Report #13) (pp. 3 - 1 8 ) . Honolulu: University of Hawai'i, Second Language Teaching & Curriculum Center.
EARLY DEVELOPMENT Think-aloud protocols and other related verbal report formats were first developed by Newell and Simon (1972) in the U S to examine problem solving activities that individuals go through to get answers to math problems (Dobrin, 1986). Newell and Simon wanted a model to make computers more efficient in problem solving. By asking people to explain the cognitive process they used to solve problems, Newell and Simon found out that people use heuristics. In other words, people apply strategies to improve performance through using feedback about what does and does not work in problem solving settings. These researchers learned that they could not teach a computer to use heuristics in the same way that human beings do (although recent neural net technology allows expert systems to more closely approximate human thinking process than has been previously possible). From this beginning, protocol analysis has evolved to identify the internal process that people use in problem solving activities, including L2 learning tasks.
USES AND TYPES A verbal report protocol is produced when a language learner verbalizes his or her thought process while completing a task or immediately after having completed a task (Ericsson and Simon, 1984; Olson et al., 1984). Cohen (1983, 1987a, 1987c) suggests three basic categories of verbal report data: self-report, self-observation, and self-revelation. Self-report is a generalized statement about one's own typical behavior. Self-observation requires that the specific processes used to accomplish a particular language task be reported simultaneously or within a very short period of time (introspection) or through probing for information soon after completing the task (retrospection). Self-revelation is a disclosure of thought processes in a stream of consciousness while the information is in the focus of the learner's attention. Selfrevelation data are basically unedited and unanalyzed; by contrast, this suggests that self-observation might be at least partially analyzed or subject to a bit of reflection. Think-aloud protocols are usually of the self-revelatory variety and are thus unanalyzed, but retrospective protocols conducted a little later than the target task might be more self-observational and thus somewhat analytic.
CRITICISMS T h e use of think-aloud protocols and other related verbal report formats is not without criticism. Tomlinson ( 1 9 8 4 ) advises researchers to be beware when retrospective protocols are gathered, because subjects may rely on their background knowledge and opinions of a topic to fill in void areas where they cannot remember what they were actually thinking. Tomlinson also points out that sometimes subjects forget that their task is to report on the mental process they have used to complete a task and not interpret or explain what they have done.
4 • ANDERSON & VANDERGRIFT
Dobrin (1986) poses four objections to think-aloud protocols and associated formats. First, he is concerned with possible methodological flaws, such as those mentioned immediately above. Second, inferences are made from protocol analyses about whole processes. These inferences are made from "traces" of mental processes. Dobrin contends that reports do not show that the traces are large enough to make inferences about the mental process that occur. Third, there is no mechanism to determine if a trace is relevant to the actual process being described. Finally, Dobrin concludes that the studies that have used protocols have not provided enough information that common sense does not already offer. Afflerbach (1986) advises that think-aloud protocols may not give helpful information if the process under investigation is automatic. "Automatic processes bypass working memory, and hence are not available for verbal reports" ( p . l ) . When the strategies or steps followed in a process have become automatic, we no longer think about what we are doing; we simply do the task.
WHICH LANGUAGE TO USE When think-aloud protocols or related formats are used with L2 learners, one consideration which is debated is which language the learner should use, the native language or the second language. Devine ( 1 9 8 7 ) recommends that think-aloud protocols to be given in the L2 learner's first language ( L I ) . She points out that "the limited oral language proficiency of many second language readers makes the use of oral summaries or retellings of text of questionable value" (Devine, 1987, p. 77). Likewise, Anderson (1989) found that L2 learners were able to provide much more information about their cognitive processes in reading if they were allowed to provide the protocol in their L I . In a later study (Anderson, 1992), advanced second language learners reported on their reading strategies in their L2 and exhibited difficulties verbally describing the strategies they had used during the reading task. However, Block (1986) showed that two E S L students, after a short orientation, were able to think aloud in English. Thus, researchers disagree somewhat as to which language is the most effective for think-alouds. T h e most reassuring situation for the student would be to allow the protocol to occur in whatever language the student is most fluent.
A POSITIVE JUDGMENT In spite of any weaknesses, the use of think-aloud protocols and their retrospective cousins provides valuable insight into a rich source of data that is inaccessible to observation and would otherwise be lost. Ericsson and Simon emphasize that for more than half a century... the verbal reports of human subjects have been thought suspect as a source of evidence about cognitive process... [Yet] verbal re-
INCREASING METACOGNITIVE AWARENESS IN THE 12 C L A S S R O O M
•
5
ports, elicited with care and interpreted with full understanding of the circumstances under which they are obtained, are a valuable and thoroughly reliable source of information about cognitive process... They describe human behavior that is as readily interpreted as any other human behavior. (1980, p. 247)
FOUR STEPS TOWARD EFFECTIVE USE OF PROTCOLS Precautionary measures can be taken to minimize any concerns about the use of think-aloud protocols and related formats. The following four steps help learners to produce useful and accurate protocols. First, the teacher or researcher provides training for learners in thinking aloud prior to having to actually produce a think-aloud protocol. Following a demonstration by the teacher or researcher, learners should be provided with an opportunity to practice. This assures everyone that the learner understands what is to be reported in the protocol. Second, the teacher or researcher elicits the protocol as close to the learners' completing the language task as possible (or even better, during the language task). This avoids putting the learner in a situation of having to rely on long-term memory of what he or she was doing during the task. Some researchers have carefully constructed the language task so that the learner is asked to provide the think-aloud protocols at specific places. For example, red dots may be placed in a reading passage every few paragraphs to signal to the readers that they should stop and verbalize their reading comprehension strategies. While gathering think-aloud protocols during written tasks, teachers or researchers inform the writers that if they are silent for more than five seconds, they will be reminded to think aloud and continue to verbalize their thought process. T h e learners providing the think-aloud protocols should clearly understand when they are to speak. Third, if the protocol is retrospective, the teacher or researcher provides the learner with some contextual information to help him or her remember thoughts that occurred or strategies that were used during the learning task. Videotaping the learners during the learning task and then allowing them to view the video while completing the think-aloud protocol is an excellent stimulus. Finally, the teacher or researcher allows the language learner to use either the LI or the L2 to produce the protocol. It is important for learners to speak freely, so they should use any language that is comfortable.
USES OF THINK-ALOUD PROTOCOLS AND RELATED FORMATS IN L2 RESEARCH Think-aloud protocols and related kinds of verbal reports have been used as part of the research methodology in L2 studies in all four language skills, i.e., listening, 6 • ANDERSON & VANDERGRIFT
speaking, reading, and writing, as well as in research related to language testing and teacher training. The use of think-aloud protocols or related formats in one research project in each of the above areas will be discussed. The main purpose of the following paragraphs is not to provide a comprehensive research review but instead to examine how verbal reports have been used to strengthen our understanding of how individuals use strategies to learn a new language. LISTENING COMPREHENSION Murphy (1987) examined the listening strategies of ESL learners. One of Murphy's purposes "was to bring listening research into the era of process-oriented research" (p. 29). The design incorporates an interesting application of the use of think-aloud protocols. Twelve intermediate ESL learners listened to six taped selections. While engaged in these listening tasks, learners were asked to "interrupt the speakers" to discuss their own thought processes. T h e interruption was signaled by a hand gesture to the researcher. Learners were encouraged to interrupt at any time they had something to say and were cautioned not to wait too long before interrupting. They were instructed to report two things during their recall protocols: a summary of what they were listening to and any other thoughts "running through their minds." Allowing the listeners the opportunity to determine when to provide the protocol left them in charge of the monitoring process. This "'stop-and-go' procedure proved to be an effective means for collecting [data]" (Murphy, 1987, p. 4). Murphy emphasizes that think-aloud protocols "provide an exploration of the listening process in action" (p. 44). SPEAKING Cohen and Olshtain (1993) have applied protocols to get L2 learners to report the ways in which they "assess, plan, and execute" their spoken utterances. T h e researchers videotaped 15 L2 learners participating in role-play situations with a native speaker of English. Six speech act situations were provided for each learner (two apologies, two complaints, and two requests). After each set of two speech acts the video was repeated for the learners, who then responded in their LI to a set of questions about what they were thinking during the role-plays. Note the strengths of this research in regards to the use of protocols. T h e videotape provided the source of the recall stimulus. T h e learners were not asked to remember great amounts of material before providing the protocol. T h e tasks were structured so that after each speech act pair the strategies for the two speech acts could be reported. READING COMPREHENSION Anderson (1991) used retrospective protocols with a group of 28 native Spanish speakers while engaged in two reading tasks: taking a standardized reading comprehension test and reading academic texts. Students gave the protocols in the LI or the L2 after reading each passage on the standardized reading comprehension test and after answering reading questions on each of two parts of two academic pas-
INCREASING METACOGNITIVE AWARENESS IN THE L2 CLASSROOM • 7
sages. By dividing the reading material into smaller chunks it was easier for the learners to recall what strategies they were using while engaged in the reading tasks. WRITING Cohen (1987b) employed written protocols to investigate how L2 learners accept teacher corrections on their written work. Through the use of a questionnaire, learners were asked to think back to the last paper on which they received feedback from their teachers (presumably a day or two before receiving the questionnaire). The survey was not administered at the same time as the papers were turned back so as not to interfere with how the learners read and interpreted the comments from their teachers. In this research the retrospective protocol was written rather than oral, providing an additional format for gathering introspective data from second language learners. LANGUAGE TESTING Anderson et al. (1991) used a special type of protocol to examine the test-taking strategies of L2 learners. To preserve the integrity of the timing of the testing condition essential to a reading comprehension test and at the same time to help the participants verbalize the strategies they were using during the exam, the testing conditions were slightly modified. The assumption was that readers might employ different strategies if allowed to take the test without time limitations. Participants were told they would have 30 minutes to complete as much of the test as possible and that after reading and answering the comprehension questions for each passage they were to say "stop." The exam time was suspended, and participants were asked to describe the reading and test-taking strategies they had used while reading the passage and answering the multiple-choice comprehension questions. Participants were to indicate when they had completed the next passage. The time was suspended a second time allowing them to describe their reading and testing strategies. The test was administered in this way until a total elapsed testing time of 30 minutes had passed. Participants were allowed to produce protocols in their LI (Spanish) or their L2 (English) or both languages. TEACHER TRAINING Johnson (1992) demonstrated that retrospective protocols can be used to enable pre-service ESL teachers to examine their instructional actions and decisions. She evaluated six pre-service ESL teachers, who viewed videotapes of lessons which they had taught. Their task while viewing the tapes was to provide a report about the decisions they were making while engaged in the act of teaching. This demonstrated most that videotapes can be an effective tool for stimulating recall.
8 • ANDERSON & VANDERGRIFT
THINK-ALOUD PROTOCOLS AND RELATED FORMATS IN THE CLASSROOM Think-aloud protocols and other verbal report formats have been developed and used primarily for research purposes. T h e following nine suggestions (which are not necessarily sequential) can be applied in adapting these research tools to the classroom (Davey, 1983; Irwin, 1991). T h e procedures below have been applied to protocols for L2 reading activities. These same basic procedures can be easily adapted for any of the other language skills. 1.
T h e teacher demonstrates the use of think-aloud protocols by selecting a passage which s/he has never read before in order to show in as natural a fashion as possible what is going on in the mind while reading. The teacher reads the passage aloud while the students follow silently. While reading, the teacher verbally reports what is going on in his or her mind. Teachers who are fluent, advanced readers, particularly native speakers, may need to slow down their thinking processes in order to be aware of what they actually do while reading.
2.
At the conclusion of the demonstration, the teacher encourages the students to add any thoughts that occurred to them during the reading.
3.
Additional demonstrations may need to be provided so that students see what is involved in producing a think-aloud protocol.
4-
Students are then grouped into pairs or groups of three and work together to practice thinking aloud. One student in the group reads aloud while the other(s) follow silently. The teacher encourages students to verbalize their thoughts and the strategies that they are using during the reading.
5.
Students who act as listeners during this activity then add their thoughts to what their classmate has already shared.
6.
T h e activity can also be done in a "reading round robin" format (Irwin, 1991). T h e class is given a reading passage, and each student is asked to read one sentence at a time and then verbalize thoughts and strategies. This activity works best if the readers cover the passage and reveal only one line of text at a time.
7.
A "hot seat" activity can also be conducted. One student reads a short passage and thinks aloud, while the others in the class follow along silently.
8.
Think-aloud protocols can also be used in regular silent reading periods. Occasionally during a silent reading activity, students are stopped and asked to verbalize what they are thinking. Alternatively, the think-aloud protocol can be implemented by having students stop at certain points and turn to a partner to verbalize their thoughts.
9.
Finally, students can practice this activity while reading silently, outside the classroom. Davey (1983) has suggested that students be asked to read silently and then complete a checklist to report the kinds of strategies they INCREASING METACOGNITIVE AWARENESS IN THE L 2 CLASSROOM •
9
were implementing during the silent reading sessions. This can very easily be conducted as a homework assignment.
A CANADIAN ILLUSTRATION OF THREE KINDS OF VERBAL REPORTS A n important study by Vandergrift (1992) in Canada used three kinds of verbal reports to assess listening comprehension of second language (French) learners. Listening involves transforming "input" into "intake," changing the buzzing whirl of spoken sound into a meaningful subset that is internalized by the learner. Listening is a complex problem-solving skill (Wipf, 1984) that is the most frequently used of all language abilities (Duker, 1971; Feyten, 1991; Oxford, 1993d). Ordinary speech contains many ungrammatical, reduced, or incomplete forms; numerous hesitations and false starts; and significant amounts of redundancy. Listening comprehension in the foreign or second language therefore requires the use of learning strategies such as comprehension monitoring, analyzing, purpose-identification, and guessing (Byrnes, 1984; Chamot et al., 1988a; 1988b; Oxford, 1990a, 1990b). Like reading, listening can be considered a skill involving both "bottom-up" and "top-down" processing (Byrnes, 1984; Richards, 1983; Rost and Ross, 1991).
OVERALL DESIGN Three different categories of verbal report, each one closer to the actual learning event, were used to elicit data on listening comprehension strategies: (1) self-report through delayed retrospection with a structured interview about strategies, (2) selfobservation through immediate retrospection after an oral proficiency interview, and (3) self-revelation through introspection with a think-aloud procedure.
SUBJECTS After permission was obtained from the school district, teachers were consulted and participants recruited from two large public high schools. School A is one of the few city schools where any "pure" language learners (those without previous French instruction) exist in their first (French 10) or second (French 20) year of French study. Students in their first, second, and fifth (French 20S) years of language study participated from School A. Since School A does not offer a nine-year French as a second language program, students in their eighth year (French 20N) were studied at School B. Participation was strictly voluntary, but only one female in French 20 declined to participate; all other selected students agreed to participate. Different numbers of students participated in each phase.
PHASE 1: SAMPLE, INSTRUMENTS, AND PROCEDURES Thirty-six students, drawn from the four course levels (10, 11, 11, and 4 students respectively), participated in the individual semi-structured interview to give retrospective self-reports of their strategy use (adapted from Zimmerman and Pons, 1986). Interviewees were asked to recall the strategies they used to comprehend 10 • ANDERSON & VANDERGRIFT
spoken French in a number of different contexts. Interviews were audio recorded, and students were encouraged to make changes and/or additions as needed to reflect accurately their strategy use. Analysis used a strategy classification scheme based on the work of O'Malley and Chamot (1990). Interrater reliability was .88 on the codings. Generally the categories of strategy analysis used in Phase 1 proved to be a useful coding guide, but a few additions and revisions were made to reflect listening comprehension strategy use and distinguish more clearly between strategies. For example, the category of inferencing was broken down into different types: linguistic (use of words in context to infer meaning), voice/paralinguistic (e.g., tone of voice, sighs), kinesic (e.g., facial expressions, body language), and extra-linguistic (e.g., background sounds). Some strategies in the original list were removed because they did not pertain to listening and others were clarified for more consistent coding. A number of affective strategies not on the list were identified. Building on the work of Oxford (1990b), these strategies were identified as lowering anxiety (use of mental techniques to make one feel more competent), self-encouragement (providing personal motivation through positive self-talk, a combination of what O'Malley and Chamot called "self-talk" and "self-reinforcement"), and taking emotional temperature (getting in touch with one's feelings and voicing them). These strategies reflect some of the ways in which learners gain control over their emotions and attitudes about learning. Another category was isolated for strategies used to deal with breakdown of comprehension while interacting with another person. Repair strategies, commonly associated with the literature on communication strategies, are important in achieving comprehension (Rost and Ross, 1991) and can be divided into direct appeals and indirect appeals for assistance (Ellis, 1986).
PHASE 2: SAMPLE, INSTRUMENTS, AND PROCEDURES Based on the Phase 1 interviews, 21 participants were chosen for stimulated recall (Phase 2). Selection was based on reported strategy use in Phase 1 and consultation with the teacher. Those reporting the greatest frequency, variety, and sophistication of strategy use were classified as more successful listeners. Those reporting the least frequency, variety, and sophistication were classified as less successful language learners. All students invited to continue with Phase 2 did so. Participants were distributed as follows: French 10 (one female successful listener, three female and one male less successful listeners); French 20 (two male and one female successful listeners, one male and one female less successful listeners); French 20S (two female and one male successful listeners, two female and two male less successful listeners); and French 20N (three female successful listeners, one male less successful listener). Phase 2 involved retrospective self-observation using a stimulated-recall session. Each participant was interviewed individually in French to determine the level of INCREASING METACOGNITIVE AWARENESS IN THE L 2 CLASSROOM •
11
proficiency (using the ACTFL/ETS Oral Proficiency interview discussed below), which resulted in a designation of either novice or intermediate proficiency. T h e videotape was played back to the participant immediately following the interview, with the participant or investigator stopping the video at any time for the participant to comment on what s/he was thinking at that point in the interview. The audiotape recorder was left on, recording both the interview and the retrospective comments. Coding of protocols for Phase 2 paid special attention to all nonverbal, indirect strategies used by the listener to clarify meaning or solicit input, as well as repair strategies. Differences in opinion about coding were resolved through discussion. Most results were useful.
PHASE 3: SAMPLE, INSTRUMENTS, AND PROCEDURES The sample used in Phase 2 also participated in Phase 3, which centered on introspective, self-revelation through a think-aloud procedure adapted from O'Malley, Chamot, and Kiipper (1989) and Rankin (1988). Students were first taught how to think aloud rather than retrospect about their listening strategies. These training sessions took 30—40 minutes. An independent group of students helped to designate various texts by difficulty level, I — IV with I being the easiest. Only authentic texts reflecting naturalness of form and appropriate situational context were used. Easier texts (I-III levels) were taken from A la radio! (Porter and Pellerin, 1989), and the most difficult texts (IV) were taken from Communication +3 (Boucher and Ladouceur, 1988). Tapes were administered to participants as follows: French 10, level I; French 20, level II; French 20S less successful learners, level III; and French 20S successful learners and French 20N, level IV. Data collection was done individually, with audio recording for later transcription and coding. Sessions (30—40 minutes each) took place within a week after training. Each session included three stages: warm-up, transition, and think-aloud. In the last stage, think-aloud data were recorded for at least three different texts, with uniform stopping points occurring for participants to tell what they were thinking.
OTHER INSTRUMENTS AND PROCEDURES T h e ACTFL/ETS Oral Proficiency Interview (Lowe, 1982) is a conversational yet structured procedure for eliciting speech samples, which are then rated for proficiency. Interrater reliability was .85 between the interviewer and an independent rater in the present study. This interview was used as part of Phase 2. T h e Learning Style Inventory (LSI) by Kolb (1985) was also applied to all participants in Phases 2 and 3. This instrument measures how people perceive reality (sensing/feeling vs. thinking) and how they process information (doing vs. watching/reflecting). Four learning style categories result when these dimensions are superimposed: (1) innovative learner or intuitors, (2) analytic learners or intellectuals, (3) common sense learners or implementors, and (4) dynamic learners or inventors.
12 • ANDERSON & VANDERGRIFT
T h e o b s e r v a t i o n s c h e m e k n o w n as C o m m u n i c a t i v e O r i e n t a t i o n of L a n g u a g e T e a c h i n g ( C O L T , A l l e n , Frohlich, a n d Spada, 1984) was used t o characterize t h e i n s t r u c t i o n a l features of t h e classes involved. T h i s is a descriptive tool to provide interactive data a n d has b e e n validated in m a n y different instructional settings. Ins t r u c t i o n a l features were coded in real t i m e o n t o a n o b s e r v a t i o n f o r m a n d n o t e s taken. PROFILES A N D O T H E R A N A L Y S E S For P h a s e 1, t h e listening profiles revealed t h e p e r c e n t a g e of d i f f e r e n t k i n d s of strategies used by e a c h individual a n d t h e average n u m b e r of strategies reported by p a r t i c i p a n t s at e a c h proficiency level. For Phase 2, a profile was created to identify t h e use a n d frequency of listening c o m p r e h e n s i o n strategies used d u r i n g participatory l i s t e n i n g at e a c h level of p r o f i c i e n c y . E a c h strategy i d e n t i f i e d d u r i n g t h e stimulated recall session (covert strategies related to t h o u g h t processes) was c o u n t e d a n d a profile created for e a c h p a r t i c i p a n t . Profiles were also c r e a t e d for e a c h proficiency level. For Phase 3, a profile was created for e a c h p a r t i c i p a n t by representing e a c h strategy and strategy group as a p e r c e n t a g e of total strategy use by t h a t particip a n t . Analysis of t h e profiles was exploratory a n d interpretive, w i t h t h e aim of hyp o t h e s i s f o r m a t i o n . S e p a r a t e analyses were m a d e for e a c h of t h e variables: (1) successful a n d less successful listeners, (2) level of language proficiency, (3) gender, and (4) learning style.
RESULTS T Y P I C A L EXERCISES In general, t h e C O L T f o u n d t h a t t h e type of listening exercise used in t h e classes f r o m w h i c h p a r t i c i p a n t s c a m e was m a i n l y d i s c r e t e - p o i n t , w i t h little e m p h a s i s o n m e a n i n g . A u t h e n t i c d o c u m e n t s were never used at t h e lower levels, a n d little global listening practice was d o n e at a n o r m a l speed. P H A S E 1: R E T R O S P E C T I V E I N T E R V I E W S Overall, t h e r e was a n increase by course level in t h e m e a n n u m b e r of distinct strategies reported (11%, 10%, 14%, 16%), e x c e p t for a slight d r o p in F r e n c h 20. Except for F r e n c h 10, females reported using more distinct strategies t h a n t h e i r male classmates (6% to 8%, 11% to 10%, 16% to 12%, 17% t o 13%). M e a n n u m b e r (and p e r c e n t a g e ) of metacognitive strategies increased at e a c h course level, a n d females reported using a greater variety of m e t a c o g n i t i v e strategies t h a n males at e a c h course level. P l a n n i n g strategies were by far t h e most popular of t h e m e t a c o g n i t i v e strategies. Cognitive strategies were t h e most widely used for all course levels, but use of these strategies appeared highest in F r e n c h 10 ( 5 5 % ) a n d t h e n decreased to 42—43%. Females reported using more cognitive strategies t h a n did males. Linguistic inferencing, kinesic inferencing, resourcing, a n d e l a b o r a t i o n were widely
INCREASING METACOGNITIVE AWARENESS IN THE L2 CLASSROOM •
13
used cognitive strategies w i t h i n t h e whole sample. T h e average n u m b e r of socio' affective strategies was small and increased at each course level, but when viewed as a percentage of total strategies reported, it rose markedly after French 10 (17%, 23%, 22%, 2 2 % ) . T h e most popular socioaffective strategies were questioning for clarification, cooperation, and self-encouragement. Repair strategies, not reported by many students, were present at advanced course levels only.
PHASES 2 AND 3: ORAL INTERVIEW FOLLOWED BY STIMULATED-RECALL RETROSPECTION AND THINK-ALOUD During t h e oral proficiency interview, it was observed that novice learners clearly relied o n kinesic inferencing, t h a t is, guessing from t h e body language of t h e speaker. Predictably, there was a gradual decline in t h e use of this strategy between novice and intermediate learners. Participants appeared to be using strategies called c o n t i n u a t i o n signals (or backchanneling cues), such as nods, " m m m m , " or faking (a n o n c o m m i t t a l response in spite of obvious n o n c o m p r e h e n s i o n , used mostly at t h e novice level). A s learners became more proficient, there was less need to seek clarification or to verify comprehension, and more advanced learners did these in t h e target language. During t h e stimulated-recall session after t h e oral proficiency interview, t h e videotape was stopped for a strategy c h e c k whenever it appeared t h e participant was h a v i n g difficulty understanding. Since strategies were elicited only w h e n participants had difficulties, strategy use decreased with each rise in proficiency level. Cognitive strategies were used t h e most to solve comprehension problems, but t h e use decreased across proficiency levels (85% to 68%). Of t h e cognitive strategies, inferencing was used at all levels, a n d transfer and translation were used by novices. Metacognitive strategies, however, increased in use across proficiency levels (2% to 32%), with comprehension monitoring and problem identification the most popular of these strategies. Socioaffective strategy use was small, appearing only a m o n g novices and including only self-encouragement. T h e s e results clearly indicate that strategy use changes across proficiency levels and also show t h a t covert listening strategies are important to assess. In Phase 3, t h e language learners were directly involved in a think-aloud procedure while listening to an oral French text. Cognitive strategies were most used by all participants (average 88%). T h e most widely used cognitive strategies were summarization, elaboration, a n d inferencing. Cognitive strategy use decreased (91% to 80%) from novice to intermediate. Novice listeners relied o n surface-processing cognitive strategies such as translation, transfer, and repetition far more t h a n did intermediate listeners. B o t h novice and intermediate listeners used elaboration, summarization, and inferencing rather o f t e n (range of 16%—30%). Overall, elaboration and inferencing appeared to be t h e most important cognitive strategies for listeners, regardless of proficiency level. Metacognitive strategies (average 12%) followed. T h e most popular metacognitive strategies were comprehension monitoring, planning including selective attention,
14 • ANDERSON & VANDERGRIFT
and problem identification, with an overall and very important metacognitive increase from novice to intermediate ( 9 % to 1 9 % ) . S i n c e t h e nature of a think-aloud interview is not conducive to eliciting the use of socioaffective strategies, this category occurred in less than 1 % of total strategy use. Few differences in actual use of strategies during the Phase 3 think-aloud occurred by gender. T h e same six cognitive strategies (summarization, elaboration, inferencing, translation, transfer, and repetition) continued to be used most often, followed by the metacognitive strategy of comprehension monitoring. A difference lay in the use of repetition, with males using it more than females ( 9 % to 7 % ) . W h i l e females did use slightly more metacognitive strategies, this difference c a n be attributed to t h e almost exclusive use of self-evaluation strategies as well as t h e greater use o f voice/paralinguistic inferencing (attributable to female sensitivity and a t t e n t i o n to metamessages, T a n n e n , 1 9 8 6 ) . O n the other hand, the greater general use of inferencing by males is n o t as easy to explain but might relate to a greater propensity for logical, rational thinking (Oxford, 1993a, 1 9 9 3 b ) . In terms of learning styles of the 21 learners on the L S I , 12 — all males except one — were identified as T y p e 2 learners (analytic, assimilating intellectuals). O f the remainder, two were identified as Type 1 (innovative, divergent intuitors); two were T y p e 3 ( c o m m o n - s e n s e , c o n v e r g e n t i m p l e m e n t o r s ) ; and t h r e e were T y p e 4 (dynamic, accommodating inventors). T w o produced poor data and were n o t included, and three more were not used because their scores fell close to dividing lines between quadrants. Sensing and feeling learners (Types 4 and 1) exhibited the same pattern of strategy use as successful learners: monitoring, elaborating in personal ways, inferring, avoiding repetition, using less non-personal elaboration, and using less transfer. T h e y generally used more metacognitive strategies t h a n did others with a different learning style. Learners who actively involved themselves in language experiences (Types 3 and 4 ) appeared to engage in more questioning and inferencing than did their more reflective counterparts (Types 1 and 2 ) , who used repetition, translation, and transfer. T h e biggest difference between successful and less successful learners seemed to lie in the use of metacognitive strategies ( 1 6 % and 8 % ) , a finding further strengthened by a qualitative analysis of selected protocols. T h e most notable differences occurred in t h e use o f comprehension monitoring and problem identification, b o t h used twice as often by successful than by less successful learners. Differences between these two groups was not as marked for cognitive strategies. Less successful learners used more transfer, inferencing, and induction/deduction, but surprisingly, translation and repetition were used almost equally by both successful and less successful learners. Even within a given proficiency level (novice or intermediate), there were differences in strategy use. Novice-low listeners relied heavily on elaboration, inferencing, and transfer, as shown by both Phases 2 and 3. T h e y were limited by cognitive constraints: the cognitive strategies needed to deal with the language did not leave enough attentional energy or memory space for deeper processing strategies such as INCREASING METACOGNITIVE AWARENESS IN THE L2 CLASSROOM •
15
comprehension monitoring. Novice-mid listeners used the above strategies but also felt compelled to translate, while novice-high listeners used superficial summarization, inferencing, and elaboration along with repetition and translation.
DISCUSSION OF CANADIAN STUDY For novices in this study, listening comprehension was an interactive process in which listeners drew upon both bottom-up (linguistic) processing and top-down (background or world knowledge) processing, with reliance on the latter to construct a meaningful interpretation of a text in memory. Novice listeners, because of their limited linguistic knowledge, relied heavily on cognates (transfer) and contextual, extra-linguistic clues such as text type, background noise, tone of voice, body language, and relationships between speakers to employ world knowledge (elaboration) and to understand the input (inferencing). The essential difference in strategy use between novice and intermediate listeners was the increased use of metacognitive strategies, primarily comprehension monitoring. This dramatic increase might have been partially explained by the difference in listening ability. Intermediate listeners have learned that success in listening and in second language learning requires metacognitive control. Those who have not learned this have probably withdrawn from language study. In spite of the increased use of metacognitive strategies, intermediate listeners still use predominantly cognitive strategies: elaboration, inferencing, and summarization (the last allowing processing of larger chunks of language). Yet because of their greater internalization of vocabulary and structures, they are able to allocate more resources to metacognitive comprehension monitoring. Inefficient surface-level cognitive strategies, such as repetition, transfer, and translation, have now been superseded in frequency by deeper processing strategies. At the intermediate level, listeners hardly ever rely on body language for meaning, hypothesis testing is done in the target language, and positive continuation signals increase four-fold; listeners can hold their own in a conversation. At this level, fewer differences in strategy use emerge between sublevels. Successful listeners appeared to use world knowledge more effectively, relating it to their own experience. Meaning was accumulated by successful listeners as the new linguistic input interacted with previous knowledge sources. Greater use of comprehension monitoring, prediction, and very rich summarization occurred. Less successful listeners experienced more problems because they squandered precious time and attentional resources on efficient surface-processing strategies and because they did not use their native-language metacognitive listening strategies. Phase 1 suggested that females reported more strategy use than males, but the findings of Phase 3 implied that gender differences in actual strategy use in a thinkaloud situation were minimal. Gender differences in actual strategy use and strategy reporting should be further investigated. Of particular value might be an exploration of the gender of the investigator (in this case male) versus the gender of the respondent.
16 • ANDERSON & VANDERGRIFT
Learning style tended to relate to language learning success. Differences between feeling/sensing learners and thinking learners were similar to those characterizing successful and less successful learners. T h e former tended to experience language learning in more concrete and global ways and more often used monitoring, personal elaboration, and inferencing — all deep-processing strategies. Analytical learners, on the other hand, focused on detail, lost the main idea, and used surfacefeature strategies such as transfer and repetition. However, because successful learners also placed at the analytic end of the continuum, it would be unjustified to conclude on the basis of this study that a particular learning style might lead to greater success in listening comprehension. This study confirms via verbal reports that listening is an active process of constructing meaning. It is not a top-down nor a bottom-up process, but an interactive process by which a listener draws upon a number of sources simultaneously (Byrnes, 1984; Rost and Ross, 1991). Listening comprehension is also influenced by external factors such as the purpose for listening, the difficulty of the text, the context of interaction, and the level of anxiety, as well as by internal student factors (Faerch and Kasper, 1986; Richards, 1983). This study confirms what LI studies have shown about the crucial role of metacognitive strategies for listening comprehension, with comprehension monitoring as a superordinate strategy that is vital to selective attention, appropriate elaborating, and successful inferencing. Given that the use of metacognitive strategies, particularly comprehension monitoring, appears to be crucial for successful learners, teaching strategies should foster the growth of metacognition among students. Prelistening activities encourage listeners to plan, organize their prior knowledge, and anticipate; teachers can foster these strategies. During the listening phase, comprehension monitoring and inferencing strategies can be developed. During postlistening, evaluation strategies can be nurtured. Authentic texts are essential (Scarcella and Oxford, 1992), along with tasks that are appropriate to students' proficiency levels (Lund, 1991) and practice time that reduces students' anxiety (Oxford, 1990b; Rubin, 1987). Using different response formats for verifying successful listening comprehension adds variety and interest (Lund, 1991; Oxford, 1990b; Richards, 1990; Ur, 1984). This study was exploratory, and suggestions arose for further research. First, a listening comprehension test should be administered to all participants at the beginning of the study, not just to those at a mid-phase in the study. Second, future research should study learner's background knowledge, linguistic knowledge, and use of planning strategies in such a way that these threads can be teased apart. Third, during the Phase 1 structured interview, the interviewer should be especially alert to self-management strategies, asking participants how they arranged conditions for successful accomplishment of listening tasks; their answers would elicit more evidence of affective strategy use. Fourth, the proficiency interview should not be used to study strategy use in interactive listening; instead, both participants should be equal partners in a communication task, in which the student is not just questioned but is required to understand the interlocutor in order to complete the task. Fifth, a larger sample at each proficiency level should be used. INCREASING METACOGNITIVE AWARENESS IN THE L 2 C L A S S R O O M •
17
Sixth, similar research at different ages should be conducted. Seventh, the effects of text difficulty on strategy use at different proficiency levels merits study. Eighth, the information processing framework used here (O'Malley and Chamot, 1990) should be reviewed. It is helpful for understanding cognitive processes but not affective processes, such as motivation, self-concept, and anxiety (Skehan, 1991). Integrated models might be even more useful. Ninth, validation of a general comprehension skill would enhance our understanding of cognitive processes in second language comprehension, for both listening and reading. Finally, strategy instruction needs to be conducted to determine its influence on strategy use.
CONCLUSIONS The use of think-aloud protocols and other verbal report formats is expanding in L2 research. As both a research tool and as a class activity, such protocols provide a window into the often hidden processes that language learners use to accomplish their purposes in the second language. Think-aloud protocols and associated types of verbal reports have been used as a research tool in studies on all four language skills, as well as for identifying test-taking strategies and teacher decision-making processes. In each case, the individual is asked to reflect upon what he or she is thinking about or doing while engaged in a task. W h e n used by students, think-aloud protocols and related formats allow learners to become more metacognitively aware of what they are doing, which can lead to improved performance in the development of language skills. T h e use of verbal reports provides an important avenue of investigation into language learning strategies. T h e Canadian research of Vandergrift demonstrates how rich and illuminating verbal report data can be for second language listening comprehension. Such data can offer keen insight into the process of language learning.
18 • ANDERSON & VANDERGRIFT
Rebecca L. Oxford University of Alabama, USA Roberta Z. Lavine University of Maryland, USA Gregory Felkins University of Alabama, USA Mary Evelyn Hollaway Mountain Brook High School, Alabama, USA Amany Saleh University of Alabama, USA
CHAPTER 2 TELLING THEIR STORIES: LANGUAGE STUDENTS USE DIARIES AND RECOLLECTION Educating the mind without educating the heart is no education at all. -Aristotle
SUMMARY This chapter highlights the value of language learning diaries and recollection for heightening learners' own awareness of their strategies as used in language tasks. Depending on the nature of the instructions given to students, diaries and recollection can also uncover affective aspects of language learning, such as anxiety and motivation. Both students and teachers benefit from diaries and recollection.
INTRODUCTION Students derive the maximum benefit from language learning strategies in developing language skills when they and their teachers are aware of and pay attention to these strategies. Strategies range from cognitive techniques such as analyzing to metacognitive techniques such as planning to social and affective strategies such as empathizing with others. Unfortunately, research tells us that teachers are often unaware of or mistaken about their students' strategy preferences (Chamot and Kiipper, 1989; Cohen, Glasman, Rosenbaum-Cohen, Ferrara, and Fine, 1979; Hosenfeld, 1977; Oxford, Lavine, and Crookall, 1989). Two useful ways to assess strategies and thus enhance teacher and student awareness are language learning diaries and recollection (sometimes called recollective studies; see Cohen and Scott, chapter 7, this volume). Both of these techniques allow students to tell their own stories. Diaries give students the chance to tell the story of their current language
Oxford, R. L., Lavine, R. Z., Felkins, G., Hollaway, M. E„ & Saleh, A . (1996). Telling their stories: Language students use diaries and recollection. In Rebecca L. Oxford (Ed.), Language learning strategies around the world: Cross'cultural perspectives. (Technical Report # 1 3 ) (pp. 1 9 34). Honolulu: University of Hawai'i, Second Language Teaching &. Curriculum Center.
learning, while recollection affords the opportunity for students to share the tale of their previous language learning. Rather than encountering skepticism or reticence, we met with students' eagerness to tell about their language learning. Our experience with both diaries and recollection is that students respond by expressing, orally or in writing, the idea that "No one ever asked me before to talk about my learning! Let me tell you all I can." These techniques — diaries and recollection — can be interpreted in either an analytical, detailed way (which enables the researcher to do quantitative calculations of strategies cited by each learner and which does not usually say much about the emotional or social situation) or in a global, narrative way (which allows qualitative impressions of strategy use in the context of many other factors, such as social setting, emotions, attitudes, and motivation). In our case, we chose to use diaries for analytic, quantitative interpretation and recollection for global, qualitative interpretation in an entirely separate investigation. We could have just as well decided to analyze diaries qualitatively and recollection quantitatively. Had we structured our research differently, we might have used both diaries and recollection in a single investigation. Our particular usage, as shown in this chapter, by no means exhausts all the possibilities. This chapter provides a research review concerning each of the two techniques, then descriptions and results of the two studies, and finally implications.
RESEARCH REVIEW ON DIARIES AND RECOLLECTION Keeping a language learning diary helps learners become more aware of their strategies at any given time and across a long period. It also helps them pay attention to specific strategies they use in certain circumstances, and it heightens their attention toward the strategies their peers use. Language learning diaries are a type of self-report which allows learners to record on a regular basis numerous aspects of their learning process, including but not limited to the use of specific language learning strategies. Often the diarist is a language student, but sometimes the diarist is a specialist who writes about his or her language learning experience. In some instances, such as the accounts by Bailey (1983) or Schmidt and Frota (1986), the diarist-learner and the researcher are one and the same. Learners can write diaries in a totally free-form manner, with no direction from the teacher as to content or format, but Rubin (1981) found that directed diary reporting was a good way to obtain learning strategy data. Sometimes diaries are kept by and for the learner alone, without being read by anyone else. However, in other cases, students share their diaries with their peers or with their teacher. When another student or the teacher actively responds to what the diarist writes, a intensely communicative dialogue occurs, creating a dialogue journal (Albertini, 1990; Maguire, 1989; Meath-Lang, 1990; Peyton, 1990; Peyton and Reed, 1990; Spack and Sadow, 1983; Staton, 1980, 1983, 1987; Steffenson and Lin, 1989). In addition, keeping a diary is an excellent means of expression for stu2 0 • OXFORD, LAVINE, FELKINS, HOLLAWAY, & SALEH
dents who are reluctant to participate orally in class (Gaies, 1983). Teachers have said that they gain great insights into their optimal instructional roles through reading diaries and participating in dialogue journals, and instructors sometimes change their way of teaching based on what they learn from diaries and journals (Meath-Lang, 1990; Vanett and Jurich, 1990). Language learners benefit by keeping a diary or journal because they become more aware of and attentive to their own learning processes. Long (1979) shows how students become "participant observers" in their own learning by writing a language learning diary. A study of the diary allows for self-evaluation, improvement, and growth by the writer (Brown, 1985b). If students share their diaries with classmates, they begin to compare their learning processes and strategies. Bailey and Ochsner (1983), Bailey (1983, 1991), Brown (1985b), and Matsumoto (1994) discuss diaries as research documents. While there are of course limitations to self-reporting formats, diaries are excellent sources for the exploration of affective, social, and cognitive variables (Jones, 1977; Rivers, 1979; Schumann and Schumann, 1977). For instance, Bailey (1983) explores anxiety and competition in diaries, Oxford, Ehrman, and Lavine (1991) and Oxford and Lavine (1992) investigate learning style conflicts between students and teachers, and Brown (1985a) examines requests for specific input by language learners. In addition, diaries can provide specific instructional information, converting these first-person reports into useful aids for assessing both the curriculum and the classroom for possible adjustments (HowellRichardson, 1989). Recollection can also be a fine source of data for heightening awareness of and attention to one's own language learning strategies (see Cohen and Scott, chapter 7, this volume; Green, Oxford, and Green, 1995). One disadvantage of recollective accounts of language learning strategies is that they are occasionally subject to memory lapses. Other potential disadvantages (which might actually be advantages in many cases) are their strong subjectivity, their personal nature, and their individuality. Despite any possible problems, recollection has powerful benefits. It can often offer a rich, multifaceted perspective. It gives the "distance of time" that can put things into perspective and allow linkages to become more apparent. Recollection allows the individual to summarize strategy use and general learning experiences over a long period of time. Unless the directions given by the researcher constrain learners to write only about certain topics in a specific and potentially quantifiable way, recollection often permits great freedom to discuss strategies, styles, emotions, attitudes, motivations, barriers, and breakthroughs. Finally, recollection is inexpensive to conduct and does not typically require the involvement of more than one person. The main requirement is a perceptive, honest, open person with clear research goals — an individual who is willing to take the time and the emotional and cognitive energy to reflect on the past. T h e next part of this chapter discusses the diary study. T h e first two authors helped plan the study and analyzed and interpreted the data; the second author conducted the entire data collection with her students.
TELLING THEIR STORIES: L A N G U A G E STUDENTS USE DIARIES A N D RECOLLECTION •
21
THE DIARY STUDY PURPOSE T h e primary purpose of this study was to use diaries to determine the learning strategies used by university foreign language students in areas of listening, grammar learning, and vocabulary learning. A secondary goal was to expand students' awareness of their own strategy use and add new strategy possibilities to their repertoire.
SAMPLE Forty-two students of Spanish ( 2 6 women and 16 m e n ) from the University of Maryland at College Park were involved in this study. T h e instructional methodology of the classes was basically communicative. T h e total sample was composed of three subsamples: 24 Spanish second-semester students in two sections containing 17 and 7 students each; and 18 Spanish fourth-semester students enrolled in an elective course. Most participants were native speakers of English, but other native languages were also represented: Farsi, French, and Vietnamese. All students but one, an English-dominant Hispanic without previous formal instruction in Spanish, were studying Spanish as a foreign language.
DATA GATHERING PROCEDURES Students were asked by the teacher, the second author, to write down how they carried out a language task or approached a particular language skill. Specifically, the teacher asked students to explain in their diaries how they approached listening comprehension, how they learned grammar, and how they mastered vocabulary. Students made diary entries in either English or Spanish several times during the semester. S o m e students chose to deal only with one skill, others with all three skills. In total, 4 0 students discussed their strategies for vocabulary, 22 wrote about their grammar strategies, and 15 explained their listening comprehension strategies.
DATA ANALYSIS PROCEDURES T h e first stage of analysis consisted of examining students' narrative diaries to determine the number and type of learning strategies mentioned. This involved conducting a c o n t e n t analysis (Krippendorf, 1980; Stempel, 1989a, 1989b) with these steps: labeling and defining strategies as described by students, who used a variety of terms; defining units of analysis; training raters to reliably identify strategies; and expanding the strategy list when necessary. In the preliminary content analyses, Oxford's ( 1 9 9 0 b ) 62-item strategy system was used. Even though it is among the most comprehensive strategy systems published to date, it was still not comprehensive enough to handle the range of specific, identifiable strategies mentioned in the diaries. Oxford's system was therefore expanded by adding new strategies, such as using sequential review, paraphrasing, and memorizing by rote, or by breaking down compound strategies into additional discrete strategies (for example, planning for a language task became two strategies, scheduling and designing a sequential plan for learning). T h e expanded strategy list included 138 strategies (available from the first
2 2 • OXFORD, LAVINE, FELKINS, HOLLAWAY, & SALEH
two authors of this chapter). Strategies shown by o t h e r research to be important for language learning were retained in this system, even if they were n o t all used by t h e students in the sample. This list was used for all subsequent c o n t e n t analyses of the diaries. Interrater reliability b e t w e e n t h e first two authors was virtually perfect (r=.995) using t h e amplified list of strategies. A further stage of analysis consisted of counting t h e frequencies of different strategies and frequencies of six general groups of strategies (i.e., cognitive, memory, compensation, metacognitive, affective, and social; see Oxford, 1990b) across all three skill areas combined (listening comprehension, grammar, and vocabulary). T h e n t h e authors calculated instances of different strategies and the six categories of strategies within each of t h e three skill areas of interest. This step proved useful, because different skills o f t e n drew u p o n strategies in contrasting ways. A n o t h e r step involved analyzing t h e frequencies of different strategies and strategy types (within a given skill) by gender. T h e most detailed level of analysis therefore examined (1) which strategies were m e n t i o n e d in t h e diaries (2) with reference to a specific skill area (3) by males and females. W e conducted chi-square tests to assess significance of differences between females and males in frequency of strategy use. T h e test-assumption was t h a t frequencies would be t h e same for b o t h genders. A n y t h i n g t h a t differed from this pattern at p C .S «. m s î; J¿ io " î< V « — tío co e s tu co -Q u - co - CO ¿2 C J£ C TÜ Im t-i vS T3
J
rvï
sec.
BS WG
~ó u
sec.
00 Pc J
OC tc u M Ç è o c
c co :
V-5 c c C/T wî" l l MU ¡2 < > ¿2 :; §C
i --C fO i «V Ü u r-l -J
DATA ANALYSIS PROCEDURES Each of the materials was analyzed according to Oxford's (1990b) strategy typology containing six categories of strategies: memory, cognitive, compensation, metacognitive, affective, and social. Absolute frequencies were determined for each strategy category in each book. All strategy frequencies were then transformed into percentages for ease of comparison across books, and means and standard deviations of these percentages were found. By percentage we mean the proportion (relative frequency) of strategies in each strategy category relative to the total strategy use per book.
RESULTS AND INTERPRETATION US ESL/EFL textbooks Results indicate that each of the six strategy categories is explicitly represented in the selected books. This is not surprising, since the authors of these books were encouraged by the publisher and by the two series editors, Rebecca Oxford and Robin Scarcella, to include strategies from all of the strategy groups. T h e total number of strategies in these four books is 344- Two books, Gill/Hartmann and Sokolik, use approximately the same number of strategies (104 and 107), with James using slightly fewer at 82. Scarcella's two chapters yield 51 strategies; therefore, at that rate it is likely that her whole book would contain over 300 strategies. (That analysis will occur later using all chapters.) The most commonly used strategy category is cognitive (122 strategies across the four books), followed by metacognitive (69), memory (52), social (40), affective (35), and compensation (26). The Gill/Hartmann listening and speaking book emphasizes cognitive strategies (54% of all strategies in that book), with metacognitive strategies coming in a weak second (13%). James' listening and speaking book emphasizes cognitive (29%) and metacognitive (23%) almost equally. In Sokolik's reading book, cognitive strategies (30%) are at almost the same proportion as James', followed by memory strategies (22%). The highest percentage of strategy use in Scarcella's two writing chapters is found for metacognitive strategies (33%), followed by cognitive (18%). Overall, cognitive and metacognitive strategy use predominated in these books. Memory, social, affective, and compensation strategies had lower mean percentages of use (respectively 14%, 12%, 11%, and 8%) across these books than did the top two groups, cognitive (24%) and metacognitive (22%). Significant differences across these books were found for five out of six of the strategy categories (memory, cognitive, compensation, metacognitive, and affective strategies), but not for social strategies.
US textbooks for Spanish as a foreign language Three U S textbooks for Spanish as a foreign language were examined: Young/Wolf, Long/Macian, and Galloway/Labarca. A total of 174 strategies appeared. Clearly, 1 2 6 • HAJER, MEESTRINGA, PARK, & OXFORD
the Galloway/Labarca book is the overall strategy leader, with 110 strategies compared with 37 for Long/Macian and 27 for Young/Wolf. In addition, the Galloway/Labarca book has substantial numbers of strategies in five of the six strategy categories (all but affective strategies). Interestingly, Galloway/Labarca and Long/Macian are integrated-skills basal texts from the same publisher, but they show distinctly different numbers of strategies. It might have been expectable that the Young/Wolf, which deals with just one skill (reading), might have only a small number of strategies (actual number was 27). However, the ESL reading book by Sokolik, mentioned above among the ESL/EFL books, has 107 strategies. Therefore, in actuality dealing with only one skill does not restrict the number of possible strategies. Cognitive strategies are by far the most popular among the six strategy groups, with a total of 68 strategies across the three books, followed by compensation strategies (35), memory strategies (34), metacognitive strategies (24), social strategies (12), and affective strategies (1) . Four of the six strategy categories (affective, memory, metacognitive, and social) have only a small number of strategies for the Long/Macian and Young/Wolf books. Percentages of strategy use in each of the six strategy categories differed for each book. Cognitive strategies lead in each book: 70% for Young/Wolf, 57% for Long/Macian, and 26% for Galloway/Labarca. However, in the Galloway/Labarca book, two other categories of strategies were almost equal in representation to cogni' tive: memory and compensation. Slightly over half the strategies in these books are cognitive, according to the cross-book mean of 51%. Far behind come the mean percentages for compensation strategies (17%), memory strategies (15%), metacognitive strategies (12%), social strategies (4%), and affective strategies (less than 1%). There were significant differences found among these foreign language books for each of the six strategy categories. This means that within each strategy type, the books used widely differing numbers of strategies.
US-UK learner guidebooks T h e total number of strategies found in the four U S - U K learner guidebooks is 264Metacognitive strategies are the most frequent (93 instances) in the four U S - U K learner guidebooks, but cognitive strategies are not too far behind (72). These are followed by social strategies (32), memory strategies (28), compensation strategies (22), and affective strategies (17). The Rubin/Thompson book provides the greatest number of strategies (106), followed at a distance by the three other books: BrownAzarowicz/Stannard/Goldin (60), Ellis/Sinclair (58), and Fuller (40). Clearly, Rubin and Thompson impart the greatest number of learning strategies to the reader. Metacognitive strategies and cognitive strategies average 36% and 29%, respectively, of all strategies across the three learner guidebooks, while the lowest categories, compensation and affective, represent only 7% and 6% of all the strategies found in
H o w PRINT MATERIALS PROVIDE STRATEGY INSTRUCTION •
127
these books. Memory and social strategies are toward the low range, with 11% and 10%, respectively. Cognitive strategies are stressed by Fuller (52.5% of his total number of strategies) and less so by Rubin/Thompson (27%), but metacognitive strategies are favored by Brown-Azarowicz/Stannard/Goldin (45%) and Ellis/Sinclair (62%). When comparing across the three learner guidebooks, significant differences are found in terms of percentages in each of the six strategy categories. Again, the books include learning strategies in very different numbers within a given category.
Dutch as a mother tongue textbooks In two textbooks of Dutch as the mother tongue, explicit strategies are rarely used, in contrast with the frequencies of strategies in the books found earlier. One of the two books (Schlebush et al.) included only one strategy, while the other (Bult et al.) included 17. T h e total for the two books was therefore 18. Schlebush et al.'s only strategy was cognitive (for verb conjugation). Bult et al., who were trying to integrate strategies from a strategic reading textbook (Lijmbach, Hacquebord, and Galema, 1991), included some strategies in each of these three categories: cognitive (in this case, skimming, scanning, using resources, note-taking, summarizing), metacognitive (planning and identifying the purpose), and compensation (guessing from context). Note that we were more tolerant about counting strategies in these books than in the U S and UK books. For instance, Bult et al. failed to label the strategies, but all other aspects of explicitness were present in that book. The mean percentage for cognitive strategies across the two books was 77%, followed by metacognitive ( 1 8 % ) and compensation (6%), with absolutely no strategies in these categories: memory, affective, and social.
Dutch as a second language textbooks Only 10 strategies are included in these two books, and nine of them are cognitive (four in Olijkan and five in Van Loon). T h e Olijkan and van Loon textbooks use cognitive reading strategies, and almost no other strategies are included. Interestingly, the authors of these two DSL books seem to use a computer-based instructional methodology common to Dutch mother tongue books (cf. Witte, 1992a, 1992b), in which the children by themselves have to discover how and when to use what they learn. Eighty percent of Olijkan's few strategies are cognitive, while the rest are compensation. In contrast, 100% of Van Loon's strategies are cognitive. T h e average percentages across the two books are 90% for cognitive and 10% for compensation, with no other strategy groups represented.
Netherlands-published foreign language textbooks (German and French) We now consider the outcomes for two Netherlands-published foreign language books, one in German (Voogt/Haelen) and the other in French (Bimmel-Esteban et al.). T h e total number of strategies in these two books is 47, and all 47 come from Bimmel-Esteban et al.; no strategies were identifiable in Voogt/Haelen. In the
1 2 8 • HAJER, MEESTRINGA, PARK, & OXFORD
Bimmel-Esteban book, the strongest category was cognitive (37 strategies, dealing with recognizing and using formulas, analyzing contrastively, and other aspects), and three other categories had a sprinkling: compensation (5), metacognitive (3), and memory (2). Almost 80% of Bimmel-Esteban et al.'s strategies (which are called "study indicators" in that book) are cognitive, followed by 11% compensation, 6% metacognitive, and 4% memory. Because of the absence of strategies in Voogt/Haelen, the means for each category are exactly half of Bimmel-Esteban et al.'s percentages: 39%, 5%, 3%, and 2% (note rounding effects). Note that the Bimmel-Esteban textbook has been adjusted to the Netherlands' national core objectives, which include strategy use. Some strategies are recycled in the exercises found in the BimmelEsteban et al. textbook.
SUMMARY AND IMPLICATIONS Before summarizing the results, we want to reemphasize that the selection of books for this study was purposive (designed for a reason, not random). Therefore, this pilot study offers only tentative answers to the research questions; more investigation must be done later on each question. In answer to Question 1, across all 17 books in general, cognitive strategies take the lead (except in the learner guidebooks). Metacognitive strategies are a popular second-place category. T h e four other strategy groups have far less representation. A total of 857 strategies are identified in the materials analyzed. W e found important differences across the types of books and across the locations, thus answering Questions 2 and 3. The inclusion of learning strategies in U S and UK language textbooks and learner guidebooks is much more frequent than that in textbooks from the Netherlands. Some recent ESL textbooks in the U S heavily use strategies in all six categories, but the range of strategy types is more restricted in U S foreign language textbooks (with a concentration on cognitive strategies and a somewhat lesser emphasis on memory, compensation, and metacognitive strategies and little interest in affective and social strategies). Metacognitive strategies are more often used in US-UK learner guidebooks than in U S language textbooks. In our sample of books from the U S (and one from the UK), far less frequently found strategy groups are memory, affective, and social. Compensation strategies vary in their representation. In the books from the Netherlands, strategy use is almost strictly confined to cognitive strategies. Other types of strategies such as memory, compensation, or metacognitive are rare, and affective and social strategies do not appear at all. Most Dutch authors have not discovered the need to include strategies in any explicit way. However, one recent book, influenced by the national core objectives, reflects a strong effort to include strategies and therefore represented more than 60% of all the strategies found in books from the Netherlands. Languages learning strategies can be linked to all four language skills (listening, reading, speaking, and writing), as shown most clearly by the U S materials, thus answering Question 4- See also Oxford (1990b) for hundreds of examples of skillrelated strategies. In the Dutch materials the sporadically used strategies are often H o w PRINT MATERIALS PROVIDE STRATEGY INSTRUCTION •
129
linked to receptive language skills, especially reading, and to subskills like vocabulary learning, although we perceive some recent tendencies toward greater languageskill variety in the inclusion of strategies. Question 5 concerns the explicitness of strategy use. The strategies analyzed above are by-and-large explicit. However, we had to be merciful in our analysis: there is a variety of ways in which the book authors work with strategies. For example, some authors of the analyzed books do not encourage student reflection on or evaluation of their own strategy use: sometimes the steps or procedures to be used in the strategy are a bit unclear, even if the strategy is clearly labeled; in one book, the strategies are not labeled; and in some instances, the benefit of using the strategies is not explained fully. In response to Question 6, we can say that student self-reflection and self-evaluation about strategy use are generally weak. O n e purpose of metacognitive strategy use is to help students evaluate their use of other strategies, but as we have seen, many books do not emphasize metacognitive strategies (and hence self-reflection or self-evaluation). W e must give only a partial answer to Question 7. Based on our non-random data, we cannot provide a full picture of strategy instruction across countries. However, we have enough data now to offer a tentative typology of strategy instruction that might be useful for further analyses. This typology is shown in Table 9.3. Although international comparisons are difficult to make, we have tried to make them in this pilot study. We want to expand our analysis to other countries in the near future and to other books within the currently selected locations. We can see the danger of strategy-weak language textbooks and learner guidebooks in which the possibilities of strategies are not exploited in their fullest and most powerful sense. W e can also envision the danger of professionals attacking the use of language learning strategies as something "new" or threatening. In the Netherlands, these trends are happening due to a lack of dissemination of research insights from abroad, compounded by the influence of conservative teachers under the Dutch educational freedom act. (Such trends are not occurring in any clear way in the U S or the UK.) Possibly the typology of strategy instruction, which shows the different ways strategies can be integrated (with Type 4 being the most useful), would help allay any fears. Table 9.3: Typology of strategy instruction Type 0: No strategies Type 1: Blind (covert) strategies
No strategies are included in the materials. Teachers can point out the usefulness of strategies in the language classroom, but the materials provide no guidance. Some strategies are integrated in traditional language teaching, not explicitly or overtly. Learners might rehearse, or they might predict the theme of a reading from the title, but they would not label the strategy, reflect on their use of this strategy, or be motivated to use it again in a different activity.
1 3 0 • HAJER, MEESTRINGA, PARK, & O X F O R D
Type 2: Some explicit strategies
Some strategies are integrated and explicitly labeled, e.g., "ANALYZING EXPRESSIONS: This strategy can help you look at linguistic forms" (followed by ordinary grammatical exercises).
Type 3: Explicit strategies with information on how to use them (explicit, procedured strategies)
Strategies are explicitly labeled. Instructions on how to use them is given. This allows students to know if they are using the strategies correctly and gives a small degree of self-control. Example: "LEARNING STRATEGY— FORMING CONCEPTS — Sometimes it may help to concentrate on grammar so you can understand unfamiliar constructions. [Instructions:] Examine the following questions. You may find their structures and/or vocabulary difficult. Rephrase each of the sentences into a simpler sentence. Then explain why you think the author chose the more complicated wording. Obviously, many different answers are possible. Compare your responses with a classmate's in order to see another possibility."
Type 4: Explicit strategies with information on how to use them, plus a challenge to reflect on and evaluate the success of the strategy (explicit, procedured, learner-tailored strategies)
Same as Type 3, except that reflection (self-evaluation) is also included. Example: "How did you start reading a text? Did it make a difference to look first at the title?"
T o strategy researchers, we emphasize the importance of indicating the types of strategies used. W e ask, if possible, for a differentiation among the six categories mentioned here or a similar set of categories. In addition, in all classroom research we must be careful to distinguish what is in the print materials from what actually happens in class. Some teachers stick closely to the materials, while others strongly diverge. Teachers can creatively supplement Type 0 materials and thus provide excellent strategy instruction (for strategy instruction without textbooks, refer to Palinscar and Brown, 1984; Helfeldt and Henk, 1 9 9 0 ) , while teachers can misuse even Type 4 materials and offer disastrous strategy instruction. In the future, we urge more development of language textbooks and learner guidebooks as agents for strategy instruction. These materials, if developed and used well, can be a strong impetus for the creation of autonomous language learners. W e also call for more research on strategy instruction that exists in current language textbooks and learner guidebooks. This study is only a beginning. Now we turn to a second investigation of print materials for strategy instruction, based on Park-Oh's ( 1 9 9 4 b , now Park) award-winning dissertation research.
H o w PRINT MATERIALS PROVIDE STRATEGY INSTRUCTION •
131
STUDY 2: PRINT MATERIALS FOR SELF-REGULATED READING STRATEGY INSTRUCTION S e c o n d language researchers h a v e m a i n t a i n e d t h a t strategy instruction c a n p r o m o t e language learning. O x f o r d et al. ( 1 9 9 0 ) reported promising results of strategy instruction based o n findings f r o m six case studies in four countries. Positive attitudes as well as increased language progress were consistently n o t e d . C h a m o t a n d R u b i n ( 1 9 9 3 ) listed n u m e r o u s studies in w h i c h strategy use was associated w i t h language a c h i e v e m e n t or proficiency. Similarly, research o n second language reading has also claimed t h a t effective reading strategies c a n be taught, thus helping learners b e c o m e b e t t e r readers (Loew, 1984; Phillips, 1984; S u t t o n , 1989). A n u m b e r of studies ( B a r n e t t , 1988a, 1988b; C a r r e l l , Pharis, a n d Liberto, 1989; H a m p - L y o n s , 1985; H o s e n f e l d , 1984; Kern, 1989) h a v e investigated t h e effects of reading strategy ins t r u c t i o n o n gains in r e a d i n g c o m p r e h e n s i o n . Based o n such research, it has b e e n suggested t h a t strategy i n s t r u c t i o n h a s beneficial effects o n reading p e r f o r m a n c e , because t h i s i n s t r u c t i o n e n a b l e s learners t o b e c o m e m o r e aware of t h e i r r e a d i n g processes a n d strategies. Self'regulated (self-directed) strategy instruction in t h e c u r r e n t study ( P a r k - O h , 1994a), c o n d u c t e d by t h e third a u t h o r of this c h a p t e r , is defined as i n s t r u c t i o n w h i c h helps readers to use a variety of effective cognitive reading strategies a n d t o raise awareness of t h e i r o w n reading processes t h r o u g h t h e p r a c t i c e of metacognitive strategies. T h i s i n s t r u c t i o n is designed t o e n c o u r a g e learners t o t a k e ultimate responsibility for their o w n l e a r n i n g a n d t o b e c o m e i n d e p e n d e n t readers w h o c a n transfer successful reading strategies t o similar r e a d i n g tasks with no direct assistance from their teachers. T h e purpose of t h e p r e s e n t s t u d y is t o i n v e s t i g a t e t h e effects of self-regulated strategy ins t r u c t i o n o n f o u r variables i n t h e r e a d i n g p e r f o r m a n c e o f university E S L students: reading c o m p r e h e n s i o n , s t r a t e g y use, r e a d i n g attitudes, a n d information-processing styles. Providing students with training in self-regulation (also referred to as self-control training) increases the likelihood of strategy maintenance and transfer — commonly acknowledged measures of the success of any kind of skill training. In the absence of this type of self-control training, learners will not become autonomous in their use of strategies and will remain dependent on their teachers, even though they may be taught to use strategies and improve their performance on specific tasks. (Wenden, 1991, p. 106) A vast body of l i t e r a t u r e in first language a c q u i s i t i o n h a s s h o w n t h a t learners' awareness of t h e i r o w n reading processes plays a significant role in improving reading c o m p r e h e n s i o n (see, a m o n g m a n y others, Baker a n d Brown, 1984; Bereiter a n d Bird, 1985). C o m p a r e d t o first language reading research, t h e r e has b e e n relatively little research investigating t h e m e t a c o g n i t i v e strategies of second language readers, w i t h exceptions being B a r n e t t (1988b) and Carrell (1989).
1 3 2 • HAJER, MEESTRINGA, PARK, & OXFORD
RESEARCH QUESTIONS 1.
How does self-regulated reading strategy instruction affect learners' reading comprehension, use of reading strategies, and attitudes toward ESL reading?
2.
What is the relationship between learning styles and changes in reading comprehension, reading strategy use, and reading attitudes after self-regulated reading strategy instruction?
3.
What are the patterns underlying the Strategy Inventory for Reading in ESL (SIRESL) (Park-Oh, 1992a) and the Attitudesfor Reading in ESL (AIRESL) (Park-Oh, 1992b)?
SAMPLE Park, the researcher, selected 64 university ESL learners enrolled in four English 120 (freshman English for nonnative speakers) sections in the spring semester of 1993 at the University of Alabama. The subjects had met the English proficiency level required by the university and had been admitted to academic programs. They represented 20 different countries, and their ages ranged from 19 to 27. T h e initial assignment of the students to each section of the course was random. T h e four sections participating in the study were randomly assigned to the experimental and control groups — two sections per group. T h e experimental group and the control group consisted of 29 and 25 students respectively.
STRATEGY INSTRUCTION USING PRINT MATERIALS Self-regulated strategy instruction in second language reading was given as the treatment to the experimental group for eight weeks. The rationale for strategy instruction was explained as part of fully "informed" instruction (Oxford, 1990b), in which learners are told all about the purpose and procedures. Print materials contained a series of cognitive reading strategies: identifying key words, identifying topic sentences, skimming, scanning, making inferences, recognizing link-words, paraphrasing, and summarizing. In addition, metacognitive strategies involving monitoring, evaluating, and planning were also included. The strategies were given to the students in a printed, take-home form as weekly learning enrichment activities that the students were asked to practice independently outside the classroom. Materials were designed to promote learner autonomy by including step-by-step practice, so the students would have little difficulty in independently completing the tasks outside the classroom. Most tasks included a trial step with instant feedback so the students could check whether they correctly understood the practice strategy. T o heighten the students' metacognitive awareness of their own strategy use, each task involved activities practicing the metacognitive strategies listed earlier. Materials were reviewed by all teachers of the four sections before the strategy instruction to determine whether there would be any problems.
H o w PRINT MATERIALS PROVIDE STRATEGY INSTRUCTION •
133
INSTRUMENTS
TOEFL reading comprehension subtest T h e reading comprehension subtest of the Test of English as a Foreign Language (TOEFL), Form 3 K T F 1 2 , was used by permission of the Educational Testing Service in the pretest and posttest to measure reading comprehension. The T O E F L reading subtest contained five reading passages on varied topics. The length of the shortest passage used in the present study was 160 words, and the longest passage was 340 words. Thirty reading items in Section 3 were adopted in the present study. Criterion-related validity of the T O E F L reading subtest used here was evidenced by a significantly positive relationship (r=.68, p